cvg/nice-slam

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

1597 stars · 209 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

NICE-SLAM is a dense RGB-D SLAM system that combines neural implicit decoders with hierarchical grid-based scene representations, published at CVPR 2022. It produces accurate dense geometry and camera tracking on large-scale indoor scenes from an RGB-D image stream.

Use cases

• reconstruct dense 3D geometry of indoor scenes from RGB-D video
• track camera trajectories in large-scale indoor environments
• run neural implicit SLAM on datasets like Replica or ScanNet
• visualize and evaluate SLAM reconstruction results interactively
• research neural implicit scene representations for SLAM

When to choose

• you need dense RGB-D SLAM with detailed geometry on large indoor scenes
• you want a research baseline for neural implicit SLAM methods
• you need camera tracking plus hierarchical neural scene representation in one system

When to avoid

• you need real-time SLAM on resource-constrained or embedded hardware
• you only have monocular RGB input without depth
• you need a production-ready, actively maintained SLAM product rather than research code

Facets

library · maturity maintenance

computer-vision machine-learning deep-learning simulation computer-vision robotics autonomous-vehicles deep-learning python slam neural-implicit-representations neural-fields 3d-reconstruction camera-tracking rgb-d research-code cvpr-2022 linux gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="cvg/nice-slam")